Techniques for Monitoring Atmospheric Pollution and Modeling

Introduction to Air Quality Monitoring and Modeling

• Monitoring Statistics and Objectives: Air quality monitoring provides impartial, reliable, and authoritative information on ambient pollution levels. However, monitoring is expensive and often provides an incomplete picture of air pollution due to its high cost and spatial limitations.

• The Role of Modeling: Modeling is an essential tool in Air Quality Management (AQM) that complements monitoring. It is particularly significant for the optimization of control strategies and must be validated by real-world monitoring data.

• Strategic Integration: Effective AQM requires an integrated approach linking monitoring with inventories of source emissions and other assessment techniques, such as interpolation and emission measurements.

• Learning Objectives:

  • Explain the impact of wind speed and direction on air quality.

  • Outline the theoretical principles and input data for AQM models.

  • Calculate atmospheric pollutant concentrations using emission, receptor, and meteorological data.

  • Interpret and present air quality information appropriately.

Families of Air Pollution Models

• Dispersion Models: Based on a detailed understanding of physical, chemical, and fluid dynamical processes in the atmosphere. They predict concentrations at specific times and locations if emissions and controlling parameters are known.

• Receptor Models: Based on relationships between a data set of measured concentrations at a receptor and a data set of emissions that might affect those concentrations.

• Stochastic Models: Based on semi-empirical statistical relationships between pollutant concentrations and various factors, regardless of atmospheric physical processes.

• Compartment or Box Models: Calculate the mean concentration within a defined volume of atmosphere by evaluating inputs to and outputs from that volume.

Fundamentals of Wind and Meteorology

• General Definition: Wind is defined as air in motion, produced by the uneven heating of the Earth’s surface by the sun. This creates varying air pressure across space and time.

• Wind as a Vector: Wind consists of magnitude (speed) and direction. It includes horizontal (West-East, North-South) and vertical (Up-Down) components.

• Formation Factors:

  • Pressure Gradient Force: The change in pressure measured across a distance. It results in a net force directed from high to low pressure. High-pressure systems (cyclones) generally move clockwise, while low-pressure systems (anticyclones) move anticlockwise in the northern hemisphere. The gradient magnitude determines wind speed according to Newton’s second law ($F = m \times a$).

  • Coriolis Force: An apparent deflection resulting from the Earth's rotation. In the Northern Hemisphere, air is deflected to the right; in the Southern Hemisphere, it is deflected to the left. The force is zero at the equator and increases with latitude and wind speed.

  • Ground Friction: Surface roughness decreases wind speed, which in turn decreases the Coriolis effect, changing wind direction as the air mass sinks in an anticyclonic manner.

Wind Speed and Pollutant Dispersion

• Dilution and Deposition: High winds result in greater dispersion and dilution as pollutants remain airborne longer. Low winds result in rapid particle deposition and higher local concentrations.

• Wind Speed Inverse Relationship: If wind speed is doubled and all other factors remain equal, the downwind concentration of a pollutant is halved.

• Measurement Units: Wind speed is measured in $m\,s^{-1}$ or knots. One knot equals one nautical mile per hour (approximately $1.15$ statute miles per hour).

• The Power Law: Friction reduces wind speed near the surface. The change with height is approximated by:

  • $u(z) = u_0 \times (z/z_0)^p$

  • Where $u(z)$ is speed at height $z$, $u_0$ is speed at anemometer height $z_0$, and $p$ is an exponent ($0.1$ to $0.4$) varying with atmospheric stability.

  • Historical Context: This law was recognized centuries ago in England for building taller chimney stacks to carry pollutants further.

Wind Direction and Local Phenomena

• Direction Conventions: Wind direction is specified by the direction from which it blows. It is reported using 16 or 32 compass points or as an angle in degrees clockwise from North.

• Land and Sea Breezes: Caused by temperature differences. During the day, land heats up, creating a pressure gradient that draws cooler air from the sea (sea breeze). The reverse occurs at night (land breeze).

• Valley and Mountain Breezes: Valley breezes occur during the day as air moves up heated valley sides. Mountain breezes occur at night as the slopes cool and air sinks downward.

Atmospheric Turbulence and Mixing

• Mechanical Turbulence: Caused by air flow over uneven ground features (trees, buildings). It is more significant in large cities. High mechanical turbulence decreases wind speed and reduces the dilution rate.

• Buoyant Turbulence: Linked to solar radiation heating the surface. It produces thermals that carry pollutants into higher wind speeds of the boundary layer. At night, downward currents occur.

• Mixing Height: The height to which a pollutant becomes vertically dispersed, determined by the interaction of mechanical and buoyancy turbulence.

Advanced Dispersion Modeling Theory

• Deterministic Models: Used for long-term planning, based on emission registers and meteorological variables. Types include:

  • Lagrangian Models: Follow a moving air mass or "fictitious particles" (e.g., tracking a neutral-density balloon).

  • Eulerian Models: Measure properties as they pass a fixed point (e.g., cup anemometers). These numerically solve the atmospheric diffusion equation.

  • Gaussian Models: A subset of Eulerian models based on the normal probability distribution. They are preferred for predicting downwind concentrations from point sources.

  • K-Models: Complicated models treating turbulent diffusion analogously to molecular diffusion.

• Spatial Scales:

  • Macro Scale: $\sim1000\,km$ (days); involves synoptic phenomena like high/low pressure areas.

  • Meso Scale: $10s$ to $100s\,km$ (hours); includes local effects like surface roughness.

  • Micro Scale: $<1\,km$ (minutes); depends on local features like urban street canyons.

The Gaussian Dispersion Equation

• The General Equation:

  • $\chi(x,y,z;H) = \frac{Q}{2\pi \sigma_y \sigma_z u} \times \exp\left(-\frac{y^2}{2\sigma_y^2}\right) \times \left[ \exp\left(-\frac{(H-z)^2}{2\sigma_z^2}\right) + \exp\left(-\frac{(H+z)^2}{2\sigma_z^2}\right) \right]$

• Variables:

  • $\chi$: Concentration ($g\,m^{-3}$).

  • $Q$: Emission rate ($g\,s^{-1}$).

  • $u$: Wind speed at release point ($m\,s^{-1}$).

  • $\sigma_y, \sigma_z$: Standard deviations of concentration distribution in crosswind and vertical directions (m).

  • $H$: Effective plume height (m).

• Specific Conditions:

  • Ground level receptor ($z=0$): $\chi(x,y,0;H) = \frac{Q}{\pi \sigma_y \sigma_z u} \times \exp\left(-\frac{y^2}{2\sigma_y^2}\right) \times \exp\left(-\frac{H^2}{2\sigma_z^2}\right)$

  • Ground level beneath centreline ($z=0, y=0$): $\chi(x,0,0;H) = \frac{Q}{\pi \sigma_y \sigma_z u} \times \exp\left(-\frac{H^2}{2\sigma_z^2}\right)$

  • Centreline at ground level from ground source ($y=0, z=0, H=0$): $\chi = \frac{Q}{\pi \sigma_y \sigma_z u}$

Pasquill Stability Classes

• Categories: Ranges from A (Strongly Unstable) to F (Moderately Stable). G is sometimes used for Very Stable.

  • A: Strongly Unstable (Strong insolation, light wind).

  • D: Neutral (Windy, cloudy, or transition; ELR = DALR).

  • F: Moderately Stable (Night, clear sky, light wind).

• Environmental Factors:

  • Cloud-free days warm the ground (instability).

  • Cloud-free nights cool the ground (stability/inversion).

  • Increased wind speed leads to neutral stability due to mechanical mixing.

  • High-pressure regions cause downward air movement, creating "anti-cyclonic gloom."

Planning and Validation Tools

• Wind Rose: Charts the proportion of time wind blows from each direction.

• Pollutant Rose: The angular distribution of pollutant concentrations relative to wind direction, used for source attribution.

• Validation via Tracers: Sulphur hexafluoride ($SF_6$) is used as a passive conservative tracer because it is stable and has no ambient background.

• Topography Effects: Valleys can concentrate flow and cause asymmetric thermal circulations (rising on sunlit slopes, sinking on cool ones). Recirculation zones in the wake of obstacles can cause pollutant buildup.

• Building Effects: Chimney plumes can be entrained into building wakes or recirculation zones, effectively reducing emission height to zero.

Recent Developments in Modeling

• Diesel Emissions: Current HGV and LGV assumptions have been found "fallacious." New "real-drive emissions" adjustments were introduced (e.g., CURED, published 2016).

• Computational Fluid Dynamics (CFD): calculates fluid movement by dividing space into cells and using Navier-Stokes equations. It handles complex topography better than Gaussian models but requires intensive computing power.

• WRF/Chem: A fully coupled meteorology and chemistry system allowing simultaneous simulation of dynamics and chemical transport without interpolation.

Receptor and Indoor Modeling

• Chemical Mass Balance (CMB):

  • $\rho = \sum_{j=1}^{p} m_j$

  • $\rho_i = \sum_{j=1}^{p} c_{ij} \times m_j$

  • Uses chemical "fingerprints" to attribute sources (e.g., lead for petrol, PAH for wood combustion).

• Indoor Air Pollution: Defined by dilution/ventilation models.

  • Steady-state concentration: $C = m / Q$.

  • With incomplete mixing: $C = m / (c \times Q)$, where $0 < c < 1.0$.

Worked Examples from Transcript

• Example 1: Burning dump, $5\,g\,s^{-1}$ $SO_2$, distance $3\,km$, overcast night, $7\,m\,s^{-1}$ wind. Source: ground level ($H=0$).

  • Method: Use Equation 5-5 with Class D statistics ($\sigma_y = 185\,m$, $\sigma_z = 65.1\,m$).

  • Result: $18.9\,\mu g\,m^{-3}$.

• Example 2: Refinery, $90\,g\,s^{-1}$ $NO_2$, height $60\,m$, distance $500\,m$, overcast winter morning, $6\,m\,s^{-1}$ wind.

  • Method: Use Equation 5-3 with Class D statistics ($\sigma_y = 36.1\,m$, $\sigma_z = 18.3\,m$).

  • Result: $33.4\,\mu g\,m^{-3}$.